GPCR Ligand Bioactivity Prediction with Physics-Informed Dual-State Query Learning

This paper addresses the challenge of predicting bioactivity profiles of small molecules against G protein-coupled receptors (GPCRs) in drug discovery. The authors propose a physics-informed dual-state query learning framework that accounts for dynamic conformational equilibria, a factor often neglected by existing deep learning methods. The approach aims to improve the distinguishability of functional efficacies by incorporating receptor conformational dynamics into the prediction model, potentially overcoming limitations imposed by the scarcity of high-resolution active-state crystal structures.